Program Evaluation in Reproductive Health: Prospective and Retrospective Applications

Abstract

This dissertation includes 2 projects: a prospective evaluation of the Residency Training Initiative in Miscarriage Management (RTI-MM) and a retrospective evaluation of a conditional cash transfer program in Mexico, Oportunidades, and pregnancy and contraceptive use among young women. We conducted a pre-test/post-test impact evaluation to assess impacts of the RTI-MM on knowledge, attitudes, and practice of office-based miscarriage management using manual vacuum aspiration (MVA). We report that the RTI-MM is independently associated with self-report management of miscarriage using MVA at follow-up compared with baseline, controlling for other factors that may influence practice, and find evidence that support staff knowledge of MVA may impact physician practice. Our qualitative process evaluation describes the implementation process, focusing on facilitators and barriers to implementation, and revealed that perceived characteristics of the innovation must be explicitly addressed in dissemination strategies, support staff should be included in practice change initiatives, and raised questions about how to best support champions and influence perceptions of the innovation. Integrating MVA into family medicine settings has implications for access to evidence-based, comprehensive care for women. In Mexico, we used several years of data (1992, 2006, 2009) to 1) describe trends in contraceptive use, education, and fertility outcomes among young women in Mexico and 2) to assess the relationship of Oportunidades conditional cash transfer program exposure with these outcomes. We developed a matched sample from the 2006 survey, used multivariable logistic regression, and calculated predicted probabilities to estimate program impacts. We find no evidence that Oportunidades reduced the probability of pregnancy among poor, rural adolescents or increased the probability of using a modern contraceptive method among women 15-24. Education was associated with an increased probability of contraceptive use and decreased probability of reporting a pregnancy among adolescents; Oportunidades may delay fertility among younger adolescents via the schooling component of the program. Health insurance had a significant impact on contraceptive use in our sample, but the overall level of contraceptive use remains very low compared to national averages. Mexico should focus on increasing the rate of contraceptive use among rural women 15-24 overall, especially nulliparous women, regardless of Oportunidades enrollment.